Key risk indicators for accident assessment conditioned on pre-crash vehicle trajectory.

Accident events are generally unexpected and occur rarely. Pre-accident risk assessment by surrogate indicators is an effective way to identify risk levels and thus boost accident prediction. Herein, the concept of Key Risk Indicator (KRI) is proposed, which assesses risk exposures using hybrid indicators. Seven metrics are shortlisted as the basic indicators in KRI, with evaluation in terms of risk behaviour, risk avoidance, and risk margin. A typical real-world chain-collision accident and its antecedent (pre-crash) road traffic movements are retrieved from surveillance video footage, and a grid remapping method is proposed for data extraction and coordinates transformation. To investigate the feasibility of each indicator in risk assessment, a temporal-spatial case-control is designed. By comparison, Time Integrated Time-to-collision (TIT) performs better in identifying pre-accident risk conditions; while Crash Potential Index (CPI) is helpful in further picking out the severest ones (the near-accident). Based on TIT and CPI, the expressions of KRIs are developed, which enable us to evaluate risk severity with three levels, as well as the likelihood. KRI-based risk assessment also reveals predictive insights about a potential accident, including at-risk vehicles, locations and time. Furthermore, straightforward thresholds are defined flexibly in KRIs, since the impact of different threshold values is found not to be very critical. For better validation, another independent real-world accident sample is examined, and the two results are in close agreement. Hierarchical indicators such as KRIs offer new insights about pre-accident risk exposures, which is helpful for accident assessment and prediction.

Safety impacts of red light cameras at signalized intersections based on cellular automata models.

OBJECTIVE: This study applies a simulation technique to evaluate the hypothesis that red light cameras (RLCs) exert important effects on accident risks. Conflict occurrences are generated by simulation and compared at intersections with and without RLCs to assess the impact of RLCs on several conflict types under various traffic conditions. METHOD: Conflict occurrences are generated through simulating vehicular interactions based on an improved cellular automata (CA) model. The CA model is calibrated and validated against field observations at approaches with and without RLCs. Simulation experiments are conducted for RLC and non-RLC intersections with different geometric layouts and traffic demands to generate conflict occurrences that are analyzed to evaluate the hypothesis that RLCs exert important effects on road safety. RESULTS: The comparison of simulated conflict occurrences show favorable safety impacts of RLCs on crossing conflicts and unfavorable impacts for rear-end conflicts during red/amber phases. Corroborative results are found from broad analysis of accident occurrence. CONCLUSIONS: RLCs are found to have a mixed effect on accident risk at signalized intersections: crossing collisions are reduced, whereas rear-end collisions may increase. The specially developed CA model is found to be a feasible safety assessment tool.

Micro-simulation of vehicle conflicts involving right-turn vehicles at signalized intersections based on cellular automata.

At intersection, vehicles coming from different directions conflict with each other. Improper geometric design and signal settings at signalized intersection will increase occurrence of conflicts between road users and results in a reduction of the safety level. This study established a cellular automata (CA) model to simulate vehicular interactions involving right-turn vehicles (as similar to left-turn vehicles in US). Through various simulation scenarios for four case cross-intersections, the relationships between conflict occurrences involving right-turn vehicles with traffic volume and right-turn movement control strategies are analyzed. Impacts of traffic volume, permissive right-turn compared to red-amber-green (RAG) arrow, shared straight-through and right-turn lane as well as signal setting are estimated from simulation results. The simulation model is found to be able to provide reasonable assessment of conflicts through comparison of existed simulation approach and observed accidents. Through the proposed approach, prediction models for occurrences and severity of vehicle conflicts can be developed for various geometric layouts and traffic control strategies.

Gap acceptance of violators at signalised pedestrian crossings.

Gap acceptance of violating pedestrians was studied at seven stretches of signalised pedestrian crossings in Singapore. The provision of the traffic light signals provide a 'safer' crossing option to these pedestrians, as compared to uncontrolled crossings and mid-block arterial roads. However, there are still people choosing to cross at the riskier period (Red Man phase). The paper discusses about the size of traffic gaps rejected and accepted by pedestrians and the behaviour of riskier pedestrians (those adapting partial gap). The likelihood of pedestrians accepting gaps between vehicular traffic as a combination of different influencing independent variables such as traffic, environmental and personal factors was studied and modelled using logistic regression.

A before-and-after study of driver stopping propensity at red light camera intersections.

This paper reports a before-and-after study which evaluated the impacts of installing and operating red light cameras at two "T" and one "X" signalized intersections on driver stopping propensity upon the onset of amber. Rather than using video cameras, a special purpose data logger working in conjunction with loop sensors was used to gather traffic parameters, vehicle stopping/crossing movements, and signal phases. Logistic modeling was employed to model the revealed stopping/crossing decisions of non-platoon vehicle drivers in response to the onset of amber with traffic, situational and behavioral variables, including their interactions. The results indicate that the variable Cam_Inst that gathered the impacts of red light camera (RLC) installation on driver decision-making at signalized intersection was significant at 5% level only for the camera approach model of the cross-intersection. The significance of Cam_Inst was undermined for the camera approaches at the T-intersections by the inclusion of a significant two-order variable defining the interaction of Cam_Inst with distance from the stop-line. One may, thus, infer that RLC has a fixed-quantum effect at the cross-intersection, but an accentuated effect with distance at the two T-intersections. Lastly, the effects of a RLC at an intersection on the stopping decision at the non-camera approaches were minimal.

A study of stopping propensity at matured red light camera T-intersections.

PROBLEM: This paper evaluates the stopping propensity of drivers at matured in-service red light camera (RLC) signalized T-intersections after a number of years of camera operation. METHOD: A special-purpose data logger working in conjunction with loop sensors was used to gather traffic parameters, vehicle stopping/crossing movements, and signal phases at camera (treatment) and noncamera (comparison) approaches. Logistic modeling was employed to determine the stopping/crossing decisions of civilian vehicle drivers in response to the onset of yellow (amber) with traffic and the associated situational and behavioral variables, including their interactions. SUMMARY: The propensity to stop at camera approaches was about 17 times more than at noncamera approaches. This finding strongly affirmed the positive effect of RLC in encouraging drivers to stop. The likelihood to stop was about 6.5 times higher for an approach that faces an opposing right-turn conflict immediately after the onset of red. Higher stopping propensity was also more prevalent during weekdays and in periods of low traffic volumes. IMPACT ON INDUSTRY: The findings provided an objective measure on the effectiveness of RLC that would be useful to evaluate intersections for camera installation as well as on the implementation and operational aspects of RLC program.

Are accidents poisson distributed? A statistical test.

The common and convenient assumption in accident count analysis, that accidents are Poisson-distributed, is reexamined. Two statistical tests, for evaluating the assumption are described and compared. It is shown that a test based upon a combinatorial analysis is much more accurate than the alternative chi-square test when accident counts are expected to be small. The more accurate test is used to reinterpret data on accident count variability, the results indicating that the Poisson distribution is appropriate for the analysis of accidents at individual sites.

Driver behaviour at horizontal curves: risk compensation and the margin of safety.

A study involving unobtrusive observation of drivers at horizontal curves before and after realignment is described. The speeds and path radii adopted by drivers in the curves before and after realignment are compared, as are the levels of side friction demanded by each driver while negotiating the curves before and after realignment. The results reveal substantial variations between drivers (with respect to speed, path radius, and side friction demand) and between the path and curve radii. While vehicle speeds increased markedly, the side friction demand was reduced for all curves except one. It is concluded that the margin of safety was increased for all curves, and this is supported by the accident data.